htmltools::tagList(rmarkdown::html_dependency_font_awesome())

# Render the R Markdown document
rmarkdown::render("portfolio.Rmd", output_file = "../portfolio.html")

Check out some examples of our visualizations


Visualization 1-

Relationship between Number of Fatalities and Time of Day

Hypothesis 1:
Null hypothesis: Drivers are just as likely to be involved in a traffic fatality at any hour of the day

Alternate hypothesis:

Drivers are more likely to be involved in a traffic fatality at specific hours during the day, namely late at night.

Explanation:

This graph supports our alternative hypothesis, and shows that early morning also is a high traffic fatality time. The speeding information is not relevant for this specific hypothesis.

Hypothesis 2:
Null hypothesis:

There is no relationship between running a red light or not and the number of fatalities in the accident.

Alternate hypothesis:

Accidents where a red light was run are more likely to have higher fatality counts.

Explanation:

Our data seems to refute the alternate hypothesis we made, as well as our null hypothesis. Accidents where a red light was not run had higher fatality counts.


Visualization 2-

Speeding and the Average Number of Fatalities from 2013-2019

Null hypothesis:

The likelihood of fatality as a result of drivers who were speeding is equal to the likelihood of fatality as a result of drivers who did not speed.

Alternate hypothesis:

The likelihood of fatality as a result of drivers who were speeding is greater than the likelihood of fatality as a result of drivers who did not speed.

Explanation:

There is not much difference and a lot of overlap with error bars between speeding and non-speeding traffic fatality occurrences. This looks in favor of the null, but we should check with more variables to see if speeding interacts with any of them.